I’m the first to admit, I’m not a sabremetric guy. Not that I don’t love stats, I really do. In fact, there’s another column kicking in my brain in which I’ll be throwing some stats out there, but mostly of the traditional kind.
I think it was Mark Twain who once said, “There are lies, damn lies and statistics.” If one really wanted to make a case, they can take all kinds of statistics to make that case. Every one of them can make perfect sense, and yet the conclusion be totally inaccurate. That’s why I often stick to the tried and true basic statistics of baseball for my enjoyment. That plus the fact that some of these new statistics require a Masters degree just to understand.
Over the past couple of years, I have started to become more knowledgeable on such new-fangled statistics such as WAR (Wins Above Replacement) and OPS (On-Base Plus Slugging Percentage), while others drift on the outskirts of my consciousness like BABIP, FIP and wOBP.
So I’m browsing through the baseball blogosphere, when I came across a story at beyondtheboxscore.com which has to do with something called the Batter Regress Tool. This is apparently a tool someone devised that uses a variety of categories from the previous three years to try to predict expected performance in the upcoming season.
According to this, the Rangers offense is set for a downgrade in 2011, the biggest downgrade of any team in the majors, as a matter of fact. According to this Batter Regress Tool, many Rangers- in particular Josh Hamilton, Nelson Cruz and Adrian Beltre- had years beyond their norms (in other words, they had more luck on their side than most) and are due to regress back to their norm in 2011.
All in all, this tool predicts a drop in run production of an astounding 39 runs. The Angels, on the other hand, are predicted to have an uptick of 11 runs in 2011. Overall, that would be a swing of 50 runs, which could indicate a much tighter AL West race.
With all due respect, I disagree with the conclusions of this tool. The problem with tools of this sort is I don’t see where injuries are taken into account. For example, on the face of it, it’s easy to predict a bit of a regression for Josh Hamilton. After all, a .359 average doesn’t come along every day and Hamilton won the batting title with a lot of room to spare. On the other hand, if this tool is using the 3-year average of 2008-2010 to predict Hamilton, remember that Hamilton missed almost half of the 2009 season with injuries, so it would be natural for this measurement to expect Hamlton to perform even lower than the slight regression I see for him in 2011.
I’m also a bit surprised Nelson Cruz is picked for as much of a regression as he is. Cruz had a very successful 2010 on the face of it, with 22 HR, 78 RBI and a .318 average. Cruz also spent three different stints on the disabled list in 2010. So, assuming a healthier year (and I’ll even grant one trip to the DL here), even if Cruz’ batting average falls in 2011, his run production is likely to improve.
Lastly, there’s Beltre. Being a new Ranger, I don’t know as much about him. I sure have heard ad nauseum about how he performs best only in his “contract years” and last year was one of those for Beltre. Here, though, I tend to take the ballpark into account. Beltre spent 2010 playing half his games in Fenway Park, a hitter’s park, after spending the year previous playing half his games at Safeco Field in Seattle, one of the majors’ pitcher-friendliest parks. Now he takes his game to another hitter’s park, Rangers Ballpark in Arlington. Thus, while Beltre may is likely to fall from the lofty .322-28-102 heights of last year, I don’t think his run production will fall as much as the Regression Tool would indicate. Hitting in between Hamilton and Cruz should help a lot too.
None of these predictions can really take injuries into account. If injuries are severe enough, sure the Rangers could score 39 less runs, maybe even more, in 2011. If everything goes right, though, I don’t really see much of a drop-off in run production this year, no matter what some statistical programs might say.